Acquiring Day-Night Wake-Sleep Rhythms

Humans are diurnal beings who are typically active during the day and quiet at night. Through eons of evolution, human physiology has come to reflect this day-night rhythm (Moore-Ede, 1986). Day-night rhythms are due to both endogenous factors (clocks inside the body) and exogenous influences (the day-night rhythms in the physical and social world) (Ikonomov et al., 1998). The internal clock lies deep in the brain in a part of the hypothalamus called the suprachiasmatic nucleus (Dickstein et al., 1998a). It receives information about what time it is outside the body, generates signals that set the timers or pacemakers of different systems, and coordinates or synchronizes those various systems.

Newborns don't seem to know day from night (Anders, 1975). Is this because their internal clocks aren't fully mature? Is it because these clocks, although fully mature, still need to be set to the light-dark, activity-rest periods of family life? Does the fact that the newborn's energy needs require her to eat in the middle of the night keep her clocks from generating robust day-night rhythms (Wright et al., 1983)? Is she working on pulling together and orchestrating the rhythms of multiple systems—temperature, blood pressure, hormone, kidney, liver—that are themselves still maturing (Price et al., 1983)? We do not know the full answers to any of these questions yet. However, this list, which could be much longer, gives some of the flavor of the complexity of the task that faces both the newborn and the family in sorting out how to differentiate day from night.

Something is known about the facet that affects parents the most: infant sleep-wakefulness (Coons and Guilleminault, 1982; Elligson and Peters, 1980; Schectman et al., 1994). Both the structure and temporal organization of sleep evolves rapidly in the first 3 to 4 months after birth (Anders, 1975; Elligson and Peters, 1980; Gehart and Maccoby, 1980). Newborns sleep a lot, typically 16-17 hours per day (Coons and Guilleminault, 1982; Hoppenbrouwers et al., 1982), and the structure of sleep-wakefulness more strongly reflects a basic 90-minute rest-activity cycle than it does a day-night rhythm (Anders, 1982; Coons and Guilleminault, 1982). Total sleep time decreases to about 14-15 hours by 3 months, and sleeping and waking bouts begin to lengthen and consolidate (Anders et al., 1992; Bernal, 1973; Coons and Guilleminault, 1982). At birth, the longest sleeps last about 4 hours, while by 3 months they can be as long as 8 to 10 hours and, for most babies, they occur during the night (Anders et al., 1992; Bernal, 1973).

Like adults, during long periods of sleep, babies drift up to brief periods of waking or near waking. Videotaped studies of infant sleep suggest that one big difference between babies who sleep through the night and those who don't is what happens during these brief waking periods (Anders et al., 1992). Babies who sleep through rouse briefly and then settle themselves back into sleep, while those who don't then cry out and waken their parents. There is some evidence that babies who fall asleep in their cribs establish patterns of self-settling, while those who fall asleep in contact with a parent signal when they wake to circumstances that are different from those when they fell asleep. Patterns of feeding also influence nighttime signaling (and thus parental sleeping through the night) (Wright et al., 1983). Bottle-fed babies lengthen the duration of their longest bout of sleep a bit faster than do breast-fed babies. While this can sound attractive to the sleep-disrupted parent, bottle-fed babies also get sick more than do breastfed babies (Beaudry et al., 1995). Because the rhythms of other systems, like growth hormone, cortisol, body temperature, and so on, are orchestrated in relation to sleep, many suspect that the development and organization of sleep helps orchestrate the day-night rhythms in these other systems (Finkelstein, 1971; Koob, 1992; Weitzman et al., 1979).

Poets and philosophers have long been intrigued by the similarities between sleep and death. Parents find these similarities more terrifying than intriguing. Many a parent has probably poked or nudged his or her soundly sleeping infant to bring about reassuring signs of life. Although controversial, some suspect that the shorter bouts of sleep characteristic of early infancy may be protective for babies whose neural systems regulating breathing, swallowing, and airway clearing are still maturing (McKenna, 1990; Trevathan and McKenna, 1994).

Concerns that very young babies will sleep so deeply that they will fail to react when air is blocked and oxygen levels get too low have led to the recent pediatric advice to avoid placing babies on their stomachs during sleep (American Academy of Pediatrics Task Force on Infant Positioning and SIDS, 1992). Similar concerns have led researchers to reexamine what happens when babies sleep alone and when they sleep with others (i.e., cosleeping) (McKenna et al., 1993, 1994). Mothers and their 3-month or older infants have been brought into sleep laboratories, where brain waves, heart rate, and respiration for both mother and infant can be monitored throughout the night. Perhaps not surprisingly, mothers and babies don't sleep as soundly when they sleep together.

Because both breast-feeding and cosleeping induce shorter bouts and less sound nighttime sleep, several researchers have speculated that they protect the very young baby from such things as sudden infant death syndrome (SIDS) (McKenna, 1990; McKenna and Mosko, 1990). This is virtually impossible to prove. Nevertheless, differences in cultural practices surrounding cosleeping and breast-feeding abound. In this context, it may be useful to think broadly about how culturally normative variations mesh with the infant's developing capacities. On one hand, it is instructive to recognize that human infants have remarkable capacities to adapt to a wide range of caregiving practices. On the other hand, understanding practices that have evolved and been sustained around the world, but are less common in the United States, can be extremely valuable. Cultural differences are not arbitrary, nor do they reflect any simple translation from beliefs into practices; rather they reflect different trade-offs among several goals that can, in turn, have differing consequences for the infant.

Learning to Regulate Crying

Crying serves to signal caregivers. Infants of many species produce calls that serve similar functions. However, only humans cry for no reason at all (paroxysmal crying) and keep crying after being picked up, fed, and otherwise apparently made comfortable (Barr, 1990). Folk wisdom says that there are hunger cries, pain cries, irritable cries, and so on. In fact, when one hears cries out of context, the likelihood of identifying the cause of the cry is very poor (Green et al., 1995). Beyond what is causing crying, caregivers usually want to know how to stop the cries and soothe the infant.

Although there are marked cultural variations in beliefs about responding to all fusses and cries, once people do respond, they follow remarkably similar soothing patterns (Barr, 1990). They say something, touch, pick up, search for sources of discomfort, and then feed. Parents run through this repertoire so often in those early weeks and months of a baby's life that they could do it in their sleep—and often do, or so it seems. Infants cry in all of the cultures that have been studied, and interestingly, crying follows a similar developmental course (Barr, 1990; Barr et al., 1987, 1996). In caregiving settings as different as the !Kung and North America, crying makes up more and more of the baby's (and caregiver's) day until it peaks, often around 6 to 8 weeks, and then begins to decline (with the timing of the peak and decline varying for different babies). But cultural variations in caregiving practices may influence the amount of crying that characterizes infants. Comparing the !Kung and North American and Northern European babies, although crying peaks at about the same time, !Kung babies cry less, even at their peak. Furthermore, it is not that !Kung babies cry less often, rather that they are soothed more quickly.

Researchers in Canada tested whether they could help North American babies settle more readily by providing them with more of a !Kung-like caregiving experience (Hunziker and Barr, 1986). They randomly assigned families to a condition in which caregivers increased by 2 hours how much they carried the baby each day. Compared with babies in the control condition (in which caregivers kept on with their regular practices), babies who were carried around more shifted their crying toward the !Kung level of the curve. They did not cry less often, but began to be soothed more quickly. Unfortunately, the rather dramatic results of this study were not replicated when something similar was tried in England (St. James-Roberts et al., 1995b). The Canadian and English studies weren't identical, and researchers are still puzzling over whether the differences (in baby slings/packs, instructions to families, control group rates of carrying, or something else) might explain the lack of replication. This failure to replicate, however, probably suggests what parents have long suspected—there is no magic bullet, except perhaps time. By 12 to 16 weeks, most babies have settled down, spend less of their days crying, seem easier to read, and much easier to soothe.

There is also some evidence that individual differences in caregiving can affect crying in the earliest months of life. Babies whose caregivers have been responsive to their distress and sensitive in reading the meaning of their somewhat ambiguous cry communications shift more smoothly into patterns of noncrying communication and spend more time in happier, less distressed states as the first year proceeds (Crockenberg, 1981). This may sound like babies have never heard of the laws of learning (i.e., reward the behavior and it should increase). However, babies who are never responded to, such as those who grow up in institutions, often cease crying almost completely by about the time they are 3 months old. Babies evidently do learn to stop crying if it is never “rewarded.” The laws of learning actually say that the most powerful way to keep a behavior going is to reward it sometimes and not others, termed “intermittent reinforcement.” The laws of learning would argue that less, but intermittently, responsive parenting should create a baby who is more fussy and whiny. This, indeed, appears to be the case. The evidence that responding consistently and readily to the infant's cries reduces crying in the long run is not as strange as it first seems.

Not all crying is the same. Some parents live in dread of having an infant with colic. Colic is a syndrome that has perplexed parents and physicians for ages. It describes a syndrome of excessive, uncontrollable crying that, following the normal crying curve, also peaks around 6 to 8 weeks and resolves between 12 and 16 weeks (with lots of variation among babies in timing and duration) (Barr et al., 1992; Karofsky, 1984; Miller and Barr, 1991). Like other crying, it also clusters in the late afternoon and evening hours. It differs from noncolic crying primarily in being difficult if not impossible to soothe. Somewhat arbitrarily, colic is often defined for research purposes by Wessel's rule of three: fussing or crying for more than 3 hrs/day, for 3 days/week, for 3 weeks (Wessel et al., 1954). How much a baby with colic cries varies from one day to the next, and from one baby to the next. Using diaries kept by parents in which they record each crying bout of 5 minutes duration or more, the crying of infants with colic has been documented to vary from 3 hrs/day to as much as 6 or more hrs/day. This compares to the 1 to 2 hrs/day of crying that parents report for babies without colic. At its worst, a baby with colic may cry most of the time she's awake, and, adding insult to injury, she seems to sleep less than babies without colic (Lehtonen et al., 1994).

The word colic refers to what has seemed to be the obvious source of this crying: something to do with the digestive system. However, despite many attempts to identify what is wrong with the baby (i.e., sensitivity to cow's milk protein, incomplete lactose absorption, gastroesophageal reflux, gastrointestinal immaturity, including allergic reactions) thoughtful reviews of the data suggest that no more than 5 percent of colic can be attributed to identifiable pathologies (Gormally and Barr, 1997). That's not much, especially considering that the rate of colic may be as high as 10 to 20 percent in Western cultures.

Even if convinced that babies with colic are not suffering from a physical disorder, scientists, like the parents, wonder about its implications for who the baby is and who she will become. By definition, infants with colic are difficult. But does this mean that they are biologically or constitutionally predisposed to have difficult, irritable temperaments once the colic has passed? The weight of the evidence suggests that this is not true (Elliot et al., 1997; Rautava et al., 1995; Stifter and Braungart, 1992), but there is still argument over this conclusion primarily because of the difficulty of measuring both crying and temperament.

It must be a relief to parents of babies with colic to find that, by about 3 to 4 months, they have an emotionally quite different baby on their hands. No longer irritable or hard to soothe, the baby's true temperament begins to be more apparent. This is true not only of infants with colic. In general, it is difficult to predict later temperament from behavior in the first 3 to 4 months of life, as infants are going through so many transitions. Stability in infant temperament appears to increase by 4 months of age.

Recently, developmental psychologists have become quite interested in identifying the roots of a temperamental pattern that is known as behavioral inhibition. Inhibited toddlers and preschoolers are very shy in social encounters (Calkins et al., 1996; Rubin et al., 1997), wary of and upset by novel stimuli, and are thought of by parents and peers as anxious and fearful (García Coll et al., 1984). The impetus for this interest is twofold. First, there is growing concern with understanding the developmental pathways that lead to later anxiety and depressive disorders. Second, neuroscience research is revealing the specific neural pathways and neurochemical processes that underlie fear responses. This work is suggesting hypotheses about why some children may be extremely anxious and fearful from an early age.

This has led to several programs of research (see Fox et al., in press; Kagan and Snidman, 1991; Kagan et al., 1998) on young children who display fearful, shy, inhibited temperaments to identify both the antecedents and consequences of these characteristic patterns of responding to people and events. Kagan and his colleagues developed a set of challenges for 4-month-olds that included highly stimulating sounds, sights, and smells that are presented to the baby in a set order. Some babies think these stimuli are great. They smile, coo, and wave their arms around as if to say, “Give me more!” At the other end of the spectrum, some babies find the stimulation to be too much. They fuss, arch their backs, cry, and struggle, as if to say, “Take them away!” When babies are tested at 2 months, how they react to these challenges doesn't predict later temperament. However, by 4 months, reliable differences emerge. Babies who react positively are more likely to become busy, active, boisterous toddlers—some would say a handful, others refer to them as exuberant. In contrast, those who find the stimulation to be too much are more likely to be fearful and shy as toddlers and preschoolers. This evidence suggests that, for some children, negative reactivity to novel stimuli as infants evolves into a shy, inhibited, anxious temperamental pattern by toddlerhood.

This is not a trivial phenomenon. About 20 percent of healthy, European-American samples display negative reactivity to novel stimuli as young infants (Kagan et al., 1998). About one-third of the reactive infants studied by Kagan and his colleagues remained highly fearful of unfamiliar events at 14 and 21 months, and 13 percent of these infants continued to show subdued and shy behavior with unfamiliar adults and peers at 4 ½ years of age (Kagan et al., 1998). At age 6, the inhibited children continued to be socially wary and reticent during their interactions with peers and an adult experimenter, and they exhibited signs of physiological stress (Kagan et al., 1987). Researchers who have focused on the socially reticent behavior of these children have found contemporaneous associations with maternal reports of both shyness and internalizing behavior problems (Coplan et al., 1994). Recently, inhibited temperamental patterns have been associated with a physiological pattern of resting right frontal EEG activation (Calkins et al., 1996; Fox et al., 1995, 1996; Schmidt and Fox, 1994), which appears to be associated with a tendency to respond to stressful events with negative affect or depressive symptomatology (Davidson, 1992). Fox and his colleagues (in press) have recently linked this physiological pattern at 9 months of age to continuity in behavioral inhibition up to age 4. We refer to the same physiological pattern when we discuss the developmental consequences of maternal depression in Chapter 9.

Once caregivers and babies have ridden the roller-coaster of rapid developmental change through the first 3 months of life and the baby's behavioral style or temperament seems to be easier to discern, the next task is to support the baby's developing abilities to regulate his or her emotions and behavior.

Although learning to sleep through the night might seem far afield from controlling outbursts of emotion, learning to wait before acting, self-monitoring, and acquiring the ability to organize segments of behavior sequentially—all of which are embraced by the term “self-regulation”—they all involve various forms of self-monitoring and response inhibition that, in turn, reflect the growing maturity of the brain, as we discussed at the beginning of this chapter. The infant's emerging ability to replace crying with other forms of communication is just the first step along a developmental progression that recruits the child's increasing competencies into more and more mature self-regulatory functioning.

Emotional Development

Compared with efforts to understand how children learn, the study of emotion in young children is relatively new. Researchers are still asking (Mascolo and Griffin, 1998): What is emotional development the development of? The answers range from the capacity to identify one's own feelings, to the development of empathy, to the ability to constructively manage strong emotions. All are correct. As more is learned, we are struck by the richness and complexity of young children's emotional lives, as well as by the remarkable accomplishments that they make in this area prior to school entry.

The ways in which researchers learn about emotional development are diverse and, in some cases, ingenious. Some have conducted fine-grained analysis of the facial expressions of young infants in interaction with their mothers; others have observed family interactions during dinnertime or bedtime (to obtain rich profiles of the family emotional climate), or engaged in conversations with young children about their understanding of emotion (often centered around hypothetical stories); still others have interviewed parents or other caregivers about the emotions they commonly observe in the children they care for. Few of these studies have followed children over more than brief periods of time, however, so knowledge of emotional development is based largely on looking separately at infants, toddlers, and preschoolers and piecing together a picture of development from these cross-sectional data. Moreover, the broad mix of approaches to studying emotional development has not been matched by attention to the variety of contexts in which emotions develop and are socialized. Unfortunately, most of what is known has been based on studies of middle-class white children, despite suggestive evidence that socioeconomic and cultural contexts confer significant variability in how children learn to interpret and express their emotions.

Emotions are biologically basic features of human functioning. They are governed by very early-developing regions of the human nervous system, including structures of the limbic system and the brainstem that have a long evolutionary history. The capacity of a newborn to exhibit distress, fear, and rage reflects the early emergence of these biologically deeply rooted emotional brain systems. Chapter 8 discusses the richly interconnected brain structures and hormonal influences that organize the arousal/activation and regulatory/recovery interplay of emotional reactions (LeDoux, 1996; Panksepp, 1998; Schore, 1994). Throughout early childhood, other brain regions (especially in the frontal neocortex) progressively mature and become interconnected with these early-developing brain regions to contribute to the development of more accurate emotion appraisals, growing capacities for emotion self-regulation, complex emotional blends, and other developing features of mature emotional experience.

Individual differences in emotion, insofar as they derive from differences in temperament, are also biologically rooted. Temperamental qualities, including a child's typical ways of reacting to events, general mood, capacities for self-regulation, and activity level, appear early in infancy and, as mentioned earlier, show signs of stability starting at 4 months. Although the ways that temperament is manifested in behavior change significantly as the child matures—the fearful child practices being brave, the highly active child learns to sit still—these qualities are biologically or constitutionally based. Some of the most distinctive temperamental features that characterize infants at birth are based on emotional response tendencies, whether they concern the baby's dominant mood, adaptability, soothability, sociability, or fearfulness of novelty (Goldsmith and Campos, 1982; Goldsmith et al., 1987). Temperamental individuality is descriptive not just of response tendencies, but also of self-regulation (Rothbart and Bates, 1998). Young children who are reticent or withdrawn in response to new or challenging situations are displaying a temperamental attribute that is both emotional in quality (i.e., fearful) and self-regulatory (i.e., inhibited), with profound implications for their social as well as biological functioning (Kagan, 1998b). Emotions are, in short, one of the most ancient and enduring features of human functioning, and they develop significantly during the first years of life in the context of social interactions and relationships.

Emotional development during the first five years offers a window into the psychological growth of the young child (Denham, 1998; Lieberman, 1993; Saarni et al., 1998; Sroufe, 1996). Newborns' emotional lives center on feeling hungry or sleepy, too cold or too hot, and other manifestations of “state,” and their emotional repertoires seem to range narrowly from crying to cooing. In contrast, physical states rarely determine the emotions of preschoolers. Their feelings hinge on how they interpret their experiences, what they think others are doing and thinking, and how others respond to them. In early infancy, emotions can be extreme and are not easily regulated by the child or, for that matter, by the parents. By the end of the preschool years, young children are capable of anticipating, talking about, and using their dawning psychological awareness of their own and others' emotions to better manage everyday emotional experience (Thompson, 1990, 1994). Their emotional repertoires have expanded dramatically and now include such feelings as pride, shame, guilt, and embarrassment that reflect developing self-understanding and social awareness. Preschoolers have also become proficient at anticipating other's emotions, adjusting their behaviors accordingly, and even hiding their emotions from others, all important social skills of which the young infant is incapable.

Emotions are, by their nature, relational (Emde, 1987, 1998). They both emerge from and provide the basis for human attachments, social communication, and prosocial as well as antisocial encounters with adults and children (Emde, 1987, 1998; Izard, 1991). A baby's cries of distress bring his caregivers to his side; his coos and gurgles illicit playful interaction. Even young infants smile in response to a smile and cry in response to a cry (Thompson, 1998a). By the end of the first year, infants are acutely sensitive to the emotional cues of other people, especially in uncertain or potentially threatening circumstances. In a process that researchers call social referencing, infants take their cues from the reassuring or anxious expression of a caregiver, which, in turn, can affect whether they continue to play comfortably or freeze in their tracks (Baldwin and Moses, 1996; Bretherton et al., 1981; Feinman, 1992; Saarni et al., 1998; Sorce and Emde, 1981; Sorce et al., 1985; Tomasello et al., 1993; Trevarthen and Hubley, 1978). By age 2, children begin to show genuine empathy towards others (Thompson, 1998a; Zahn-Waxler and Radke-Yarrow, 1990). They not only read and adjust their own responses to others' emotions; they try to make others, even their dolls and stuffed animals, feel better. In the years that follow, a child's emotional life is shaped by relational influences as diverse as the security of attachment relationships (Cassidy, 1994; Laible and Thompson, 1998), parent-child conversations about emotional events (Kontos et al., 1994), and parental coaching of appropriate emotional expressions in social situations (Miller and Sperry, 1987). Emotional development is thus a window into how, early in life, developmental change and responsive relationships are entwined.

The cultural meanings expressed in these relationships also profoundly affect how children learn to construe and react to their emotional experiences (Eisenberg, 1986; Miller, 1994; Miller et al., 1996; Ochs, 1986). Cultural values affect how young children learn to interpret and express their experiences of fear, anger, shame, pride, embarrassment, and other emotions. They also guide the construction of new emotions and emotional blends (such as fear-shame, anger-guilt) that color emotional life and reflect these values. The sociocultural context also guides how emotions are socialized so that, for example, in some contexts experiences like teasing can serve constructive purposes and in other contexts they can be debilitating to the socialization of emotion and its expression (Briggs, 1992; Corsaro and Miller, 1992; Eisenberg, 1986; Miller and Sperry, 1987). As we discuss in the context of language development (see Chapter 6), emotions are also socialized in the context of parent-child discourse, as well as in the conversations that young children overhear among the adults around them. Chinese and Chinese-American mothers, for example, are more likely than European-American mothers in the United States to emphasize moralistic themes and the shame inherent in misbehavior when recounting their children's mischievous behavior to other mothers in the child's presence (Miller et al., 1990, 1996). In each case, mothers are interpreting the child's experience consistently with the broader values of their culture.

Because of the diverse and powerful ways that parents socialize emotional development, researchers have devoted considerable attention to the broader emotional climate of the home environment and its impact on young children (e.g., Gottman et al., 1997). In typical circumstances, young children are faced with a variety of emotional demands at home and are assisted in understanding and managing them by the support of their caregivers. When the emotional climate of the home is undermined—such as when parents experience marital dysfunction (Cummings, 1987; Cummings and Davies, 1994a; Davies and Cummings, 1994; Grych and Fincham, 1990; Lieberman and Van Horn, 1998), or parent-child interactions are conflictual, coercive, or abusive (Gaensbauer and Sands, 1979; Patterson et al., 1989, 1992), or the parent suffers from depression or other affective disorders (Dawson et al., 1994; Garber et al., 1991; Zahn-Waxler and Kochanska, 1990; Zahn-Waxler et al., 1991)—young children are confronted with conflicting, confusing, and sometimes overwhelming emotional demands. To make matters worse, they are often deprived of the parent as a resource for managing these powerful emotions (Thompson and Calkins, 1996; Thompson et al., 1995). It is not surprising to find that children from homes with serious marital conflict, or who are raised by a depressed parent, or who are maltreated are more likely than children who are spared these experiences to experience difficulties in emotion regulation and, for a small but significant minority, to develop affective disorders of their own.

What has not been appreciated until recently is that these disorders can be apparent early in life, as emotional experience is becoming organized in infancy and early childhood (Emde et al., 1993). Young children growing up with a parent with an affective disorder are themselves at risk for depression and other affective disturbances, partly as a consequence of the child's overinvolvement with the parent's emotional state and difficulties in managing the emotional demands this presents (Zahn-Waxler and Kochanska, 1990; Zahn-Waxler et al., 1988). Some young children (especially those who are dispositionally fearful or inhibited) are prone to anxiety-related disorders, especially when their temperamental vulnerability is coupled with anxious or hostile attachments to their caregivers (Cassidy, 1995; Thompson, in press(b); Vasey, 1998). Young children who are abused or neglected by their caregivers can experience significant emotional and psychosocial problems, including the display of intense, maladaptive emotions, difficulties in understanding emotion in others, and social incompetence (e.g., Beeghly and Cicchetti, 1994; Cicchetti, 1990; Rogosch et al., 1995; Shields et al., 1994). Seriously problematic parent-child interactions are also one of the cluster of features that distinguish young children who display conduct problems (Shaw et al., 1994, 1996).

These and other illustrations of the development of emotion dysfunction in early childhood suggest that, as a result of their reliance on the emotional support of their caregivers in understanding, experiencing, and managing their own feelings, young children may be particularly vulnerable to emotion-linked disorders when parent-child relationships are insecure, coercive, or otherwise troubled. This is especially so because of limitations in their independent capacities for emotional and behavioral self-regulation (Cole et al., 1994). Not only are these children failing to receive the positive supports they require for healthy emotional development; they are grappling, sometimes on a daily basis, with circumstances that undermine the well-being of many adults.

Learning to Understand Emotions

How do the indiscriminate cries of the infant become the preschoolers' differentiated feelings of anger, shame, embarrassment, and loss? Researchers have made some inroads into understanding how the generalized arousal conditions of early infancy mature into the more refined emotional states apparent in older children and adults. Social experiences are a central part of the answer (Saarni et al., 1998). In practice, a child's emotional repertoire is socially constructed.

As parents and other caregivers respond to an infant's emotional expressions, manage the child's feelings, and later label and discuss emotional experience, they help to organize and give meaning to early emotional experience. For example, parents who discuss emotions more frequently and elaborate on emotional experiences (e.g., Why do you think Sally is sad? Do you think she misses her sister? What do you think you could do to cheer her up?) tend to have children with more accurate and elaborated understandings of emotion (Brown and Dunn, 1996; Brown et al., 1996; Denham, 1998; Denham et al., 1994; Dunn, 1994; Dunn et al., 1991; Nelson, 1993, 1996). Research on children with developmental disabilities indicates that such conversations are crucially important. However, looking specifically at children with Down syndrome, their mothers are significantly less likely to refer to inner states (feelings and cognitive states) during everyday conversations compared with mothers of normally developing children, and the children are significantly less likely to talk spontaneously about feelings and more likely to have delayed expressive language skills (Beeghly and Cicchetti, 1997; Tingley et al., 1994). Parents also help their children understand that “how I feel” is not necessarily the same as “how you feel” and thus foster a growing awareness of other people as their own actors—an essential building block of social understanding. Parental behavior can also serve as a catalyst for early empathic responding, especially when parents fix their children's attention on another's distress and explain the causes of that person's feelings (Zahn-Waxler et al., 1992).

Emotion understanding grows in concert with the development of other forms of knowledge and learning in early childhood. Emotions are objects of children's thinking as well as of their feelings. Take, for example, the burgeoning research on “theory of mind,” which focuses on children's developing frameworks for inferring what other people (or children) are thinking, intending, believing, and feeling and then making predictions about how they will respond (Astington, 1993; Bartsch and Wellman, 1995; Flavell and Miller, 1998). Young children, it turns out, are remarkably perceptive. Even 2-year-olds, for example, understand that people have inner experiences of perceiving, feeling, and desiring and that they will feel good if they get what they want and feel bad if they don't. Four- to 5-year-olds appreciate the more complex connections between emotions and an individual's thoughts, beliefs, and expectations (Bartsch and Wellman, 1995; Flavell and Miller, 1998; Wellman, 1990; Wellman et al., 1995). They understand, for example, that individual tastes and preferences guide how people respond emotionally to rock music or a symphony. Their appreciation of the connection between emotions and expectations appears also in their delight in fooling others—such as telling a sibling that the thermos contains hot chocolate, when water is really inside.

In the years that follow, children begin to understand the dynamics of emotional experience, including how emotional intensity gradually dissipates over time, how specific emotions are related to certain prior events, and how a person's background, experiences, and personality can yield unique emotional reactions to events (Gnepp and Chilamkurti, 1988; Harris et al., 1985; Thompson, 1989). Eventually children also begin to understand how a person can experience simultaneously multiple or conflicting emotions, but this is a conceptual achievement beyond the grasp of most preschoolers. These are momentous achievements, but the more important point is that children's emotional and cognitive development support each other. Certain emotional understandings and capacities (including regulatory capacities) are simply beyond their reach until they have reached certain levels of cognitive maturity. The combination of emotional difficulties and communication problems (e.g., major speech delays, hearing problems) in young children provides a more troubling example of how different facets of development interact to facilitate or undermine their overall functioning (Prizant et al., 1993).

Understanding emotion is also closely linked to the growth of self-understanding and, with it, the capacity to experience self-conscious emotions such as pride and shame. Late in the second year and during the third year, young children begin to call themselves by name, to use “I” and “mine,” and to assert their new feelings of competence and independence by insisting on “doing it myself” (Bates, 1990; Bullock and Lutkenhaus, 1988, 1990; Stipek et al., 1990). Along with this newly emergent self-consciousness, young children begin to exhibit pride, guilt, shame, embarrassment, and other self-referent emotions (Barrett et al., 1993; Lewis, 1993; Tangney and Fischer, 1995). Once they have a self, in other words, young children begin to respond to experiences—good and bad—on behalf of the self. Catching a ball prompts not only glee, but also pride; spilling food in one's lap brings not only discomfort, but also shame and embarrassment. Interestingly, this is also the beginning of self-regulation and self-control (Kopp and Wyer, 1994). These self-conscious emotions are advanced by the young child's capacity to see the self as an object of observation and evaluation by others, to understand behavioral standards, and to apply those standards to an evaluation of their actions (Kochanska and Thompson, 1997; Kochanska et al., 1995). Guilt derives, for example, from the realization of how one's behavior has departed from an accepted standard, while pride arises from awareness of how one's actions have exceeded what might reasonably have been expected. With these transitions in self-awareness, the young child is an emotionally much different person than before. Her emotional life is now colored by the feelings evoked by others' evaluations and by her own self-judgments. Emotional development thus depends on and affects the development of the self.

Learning to Regulate Emotions

As young children acquire a better understanding of emotions, they become more capable of managing their feelings (Fox, 1994; Garber and Dodge, 1991; Kopp, 1989; Thompson, 1990, 1994). Emotion regulation is perhaps the most challenging aspect of emotional development. Regulating feelings depends on putting understandings about emotion to work in real-life contexts that can be extremely frustrating, upsetting, or embarrassing. Even positive emotions require regulation; exuberance is appropriate on the playground but not in a hospital. The task of emotion regulation is not simply a matter of learning to suppress emotions. It is more broadly one of deploying emotions effectively in relationships, while playing and learning, and in a wide range of settings. The fact that cultures and even individual families differ in their standards for the appropriate display and management of emotions makes this task especially challenging for children who cross over from one cultural environment (e.g., their immigrant family) to another (e.g., their preschool, which may or may not reflect their home culture).

Children start life completely incapable of modulating the expression of overwhelming feelings, integrating emotions adaptively into the fabric of social interactions, and deploying emotions in the service of focusing and sustaining attention. From a very early age, however, infants develop rudimentary skills for managing their own emotional experiences, in part by learning to enlist others who can help them. This can be observed initially in the comfort seeking of a distressed infant or toddler (Thompson, 1990). By the middle of the second year, toddlers can already be observed making active efforts to avoid or ignore emotionally arousing situations, engaging in encouraging or reassuring self-talk, changing or substituting goals that have been frustrated, and other quite sophisticated behavior strategies for managing emotions (Braungart and Stifter, 1991; Bretherton et al., 1986; Buss and Goldsmith, 1998; Calkins and Johnson, 1998; Cummings, 1987; Grolnick et al., 1996; Smolek and Weinraub, 1979; Stein and Levine, 1989, 1990). By the time of school entry, children's regulatory repertoires have become increasingly proficient and flexible as they learn, for example, that their interpretations of events can affect how they react and that they can camouflage their emotions if need be (Harris, 1993).

Children who learn to manage their emotions constructively not only have an easier time with the disappointments, frustrations, and hurt feelings that are so prevalent when they are young—and are presumably happier as a result—but they also have an easier time relating to others at home, in child care, and on the playground, as we discuss in more detail in Chapter 7. Indeed, the close correspondence between emotion regulation and relationships with peers has critical implications for efforts to foster positive social interactions and help young children who have problems in this area. Acquiring the capacity to regulate emotions also helps children believe that emotions are manageable, controllable, and can be appropriately mobilized and expressed—in short, that one's feelings need not be overwhelming, undermining, or disorganizing —what Saarni (1990, 1999) calls “emotional self-efficacy.” Children who do not feel in control of their emotions are more prone to outbursts, inattention, and rapid retreats from stressful situations, thereby creating a self-fulfilling prophecy (Garber et al., 1991). Finally, the capacity for self-regulation is a prerequisite for the critical task of learning to comply with both external and internalized standards of conduct (Zahn-Waxler and Radke-Yarrow, 1990; Zahn-Waxler et al., 1992). Compliance is dependent on the child's ability to control his reactions, as well as his motivation to do so.

Parents contribute in multiple ways to children's developing capacity for emotion regulation. As with the infant's emerging ability to regulate states like crying, the parents' role in socializing emotion regulation is one of gradually handing over the reins to the child. At first, parents and other caregivers intervene directly to soothe or pacify the infant, organizing the child's experiences around routines that are manageable and predictable. These behaviors lend predictability to the infant's world, reduce the emotional demands of daily experiences, and “scaffold” the infant's own efforts at emotion regulation. Later, parents and others coach children in strategies for mobilizing their emotions to fit the needs of a given situation, whether it involves comforting a hurt friend, learning to take turns, or dealing with the frustration of attempting a task that is just beyond their capabilities (Thompson, 1990). This involves a subtle blend of give-and-take with the child (i.e., you hide and learn that you can handle the anxiety of doing without me momentarily, and then I'll find you and dispel your anxiety), the provision of supportive challenges (i.e., this puzzle may be a little frustrating, but I'll help you do it), and respect for the child's unique ways of dealing with emotions (i.e., I'll let you blow off some steam before I try to calm you down). Equally important, however, are the more subtle ways in which the reassurance that young children derive from their attachments to caregivers constitute an important resource for emotion regulation (Cassidy, 1994, 1995; Cassidy and Berlin, 1994; Nachmias et al., 1996). Emotion regulation is fostered, in other words, not only by the parent's immediate interventions but also by the security and confidence that the relationship with the caregiver inspires in children as they grapple daily with feelings that, initially without even a vocabulary to describe them let alone strategies for managing them, can be confusing and frightening (Case, 1992; Diamond, 1996; Diamond and Taylor, 1996; Diamond et al., 1994; Johnson, M.H., 1998; Rothbart et al., 1990, 1994; Stuss, 1992).

Parents, however, do not operate in a vacuum. Research is revealing the large extent to which the task of learning how to manage one's emotions and integrate them into daily life is a different challenge for children with different temperaments, as well as for their parents. The ability to inhibit a response one is all set to perform, sometimes called effortful control, has been of special interest to researchers who seek to understand how individual differences in children's tendencies to respond to stressful or exciting events affect the growth of emotion regulation. Effortful control is one component of a larger set of inhibitory competencies, termed “executive functions,” discussed later in this chapter.

Effortful control is what enables the preschooler to take a response that is “primed” and inhibit it. Examples include acting only when it is appropriate to the rules (such as when the game leader says “Green light!”), constraining negative emotional outbursts, and planning a long-term strategy for a desired reward (such as saving an allowance to buy a Nintendo game). A game like Red Light, Green Light or Simon Says, for example, involves getting one type of behavior going (like walking quickly to the finish line or doing whatever Simon says as fast as you can) and then suddenly inhibiting or stopping those actions. Doing well at these games requires paying sharp attention for a long time, keeping track of the rules of the game, and interrupting actions that you are all set to perform. The more excited children get about playing these games, the harder it is for them to play them well. The faster they get going, the harder it is to stop. In the language used at the beginning of this chapter, the stronger the reaction, the greater the challenge for regulation.

The ability to play these games (and to use these competencies in general) seems to involve the development of structures in the prefrontal areas of the brain (Rothbart et al., 1995). An area of the frontal lobe called the anterior cingulate gyrus becomes very active any time one effortfully attempts to inhibit a thought or action for which one is primed. This area of the brain, like many areas in the frontal lobe, develops slowly over the course of childhood. People get better at performing the cognitive tasks that involve the anterior cingulate as they get older, and they also get better at controlling their emotional expressions as they get older. What researchers are now trying to determine is whether these two phenomena go together. Children of the same age exhibit differences in behaviors that should reflect the development of the frontal lobe. Parents and teachers report that some children are better than others at inhibiting inappropriate behavior, playing games like Red Light, Green Light and so on. So far, for normally developing children, the most evidence that effortful control is involved in emotion regulation comes from studies using parent and teacher questionnaires and observational tasks (i.e., not peeking while an experimenter wraps a present for you) involving behaviors that should reflect these competencies. There is very little evidence as yet that relates data obtained from these methods to activity in the anterior cingulate or other areas in the frontal lobes. Thus, the link to brain development is still only a theoretical one.

Children develop effortful control competencies gradually over the preschool years, and the full expression of these competencies requires development that extends into adolescence. From early in their development, some children seem to be better at effortful control than others, and there appears to be reasonable stability in this aspect of temperament and regulatory capacity (Kochanska et al., 2000). There is also growing evidence that individual differences in these capacities have meaningful implications for several aspects of early development that parents and others who work with young children care a great deal about. For example, young children who are higher on measures of effortful control tend to perform better on measures of early conscience and moral behavior (Kochanska et al., 1996, 1997). Conversely, infants and young children who have difficulties with inhibiting more compelling, negative impulses also tend to elicit aversive responses from others which, in turn, recreate precisely the kinds of experiences that lead to impulsive and negative behaviors (Rothbart and Bates, 1998). Not surprisingly, children who are not good at effortful control have a hard time with peer relations (see Chapter 7). In each of these examples, the response biases that come with a young child's unique temperamental profile provide the intrinsic context within which developing capacities for self-regulation emerge.

In sum, self-regulatory skills have important implications for how well children negotiate many other tasks of early childhood. Identifying and intervening with children who need extra help in developing these competencies may be important. However, determining who really needs help, as opposed to just more time to grow up, may be difficult. Furthermore, it seems possible that children who have more to regulate (i.e., those who are more exuberant and more active, more anxious and inhibited) may appear to be delayed or deficient in self-regulatory abilities, when in fact they are not. They may simply need to reach more mature levels of these abilities to be able to adequately manage who they are.

Emerging Capacities for Executive Functioning

Early researchers did not study executive functioning in young children, believing that executive skills were not functional until the brain reached maturity in adolescence (Golden, 1981). It is now generally recognized that early precursors of these skills are present in infancy (Welsh and Pennington, 1988), and there is a growing body of research that demonstrates that performance on executive tasks improves in a stage-like manner that coincides with growth spurts in frontal lobe development during infancy and through the early childhood years (Anderson, 1998; Bell and Fox, 1992, 1994; Levin et al., 1991; Posner et al., 1998; Thatcher, 1991; Welsh and Pennington, 1988). This evidence for the early emergence of executive skills is further supported by findings from the neuropsychological literature that link deficits in executive function to early frontal lobe dysfunction (Benton, 1991; Eslinger and Grattan, 1991; Tranel et al., 1994).

What are the first signs of emerging executive skills, and when do they develop? At a very basic level, executive functions cannot emerge before the child is able to orient to relevant and important features in the environment, anticipate events, and represent the world symbolically (Barkley, 1996; Borkowski and Burke, 1996; Denckla, 1996; Pennington et al., 1996). Recent methodological advances have made it possible to study some elements of these abilities in infants. For example, Haith and his colleagues have demonstrated that infants as young as 6 weeks are capable of anticipating a sequence of events (Dougherty and Haith, 1997; Haith and McCarty, 1990; Haith et al., 1988). When they are shown pictures that appear and disappear in predictable locations at predictable times, 6-week-old infants quickly form expectations and demonstrate they can anticipate the location of the next picture by shifting their eyes to the predicted location before the picture appears (Haith et al., 1988). The ability to make anticipatory eye movements using these simple sequences improves with age, becoming consistent by 3 1/2 to 4 months (Haith et al., 1988; Johnson et al., 1991), but then it reaches a plateau in which there is no improvement between 4 and 10 months (Posner et al., 1997). It is not until age 18 months that infants can anticipate ambiguous, context-dependent sequences (e.g., learning to look at a target that moved from location 1 to location 2 and then back to locatoin 1 before moving to location 3) that would require focused attention in adults (Posner et al., 1997). These emerging abilities to control attention underlie the development of executive functions that entail, for example, planning and executing sequences of behavior.

Means-ends behavior, another precursor to executive functioning, commonly emerges around ages 8 to 12 months, when the infant will, for example, remove an obstacle to retrieve a toy (Piaget, 1952). Research aimed at linking the emergence of goal-directed behavior to early brain development has provided evidence that frontally mediated, goal-directed, planful behavior is present as early as 12 months in infants (Diamond, 1988; Diamond and Goldman-Rakic, 1989; Goldman-Rakic, 1987). At about the same time, children learn to use language and to represent the world through symbols. Symbolic representation and language are the means through which a child can link the present with past knowledge and a future goal (e.g., Baron and Gioia, 1998; Kopp, 1997). They are believed to be the cornerstone for working memory and a necessary component for executive problem solving (Goldman-Rakic, 1987).

A third skill that emerges in infancy and continues to develop through childhood is self-control (Kopp, 1982). Research on self-control examines the child's emerging ability to comply with a request, to inhibit or delay an activity, and to monitor behavior according to the situational demands (Kopp, 1982). The ability to exercise self-control increases from 18 to 30 months and becomes more stable across time and across situations (Vaughn et al., 1984).

The capacity to use developing executive function to regulate behavior and emotions in the service of social goals and situational demands is sometimes referred to as inhibitory or effortful control, as discussed above. Because many skills, competencies, and experiences affect whether a child can regulate his or her emotions and behavior, researchers have used a wide variety of tasks to assess individual differences in effortful control. These tasks include being able to shift with ease from doing something as “fast as you can” to “as slow as you can” to being able to “not peek” when waiting for a surprise gift, to being able to play games like Simons Says. When individual differences on such tasks are assessed and averaged, they provide one window into why some children comply more readily with adult requests not to touch interesting things even when the parent is not watching and more readily resist the temptation to cheat on games even when they think they will not be caught. Being good at effortful control tasks, including those that more directly assess executive functioning, doesn't mean that a child will behave in compliance with social rules that require self-control, however. Aspects of children's relations with others that motivate them to want to adopt the rules of their group also matter (Kochanska, 1990).

A number of researchers have investigated the developmental trajectories of executive function by presenting children with a battery of tests purported to measure different aspects of this domain of regulatory behavior (e.g., Gnys and Willis, 1991; Levin et al., 1991; Welsh et al., 1991). The focus here is not on precursors of executive function, but on manifestations of behaviors that constitute components of this construct. These studies have demonstrated that the different component skills involved in executive functioning show different developmental trajectories and mature at different rates. In one of the first studies to include preschoolers, children ages 3 to 12 were presented with a series of tasks that involved visual searching, verbal fluency, motor planning, planning sequences, the ability to respond flexibly to changes in the environment, and the capacity to inhibit responses (Welsh et al., 1991). Patterns of performance on these measures indicated that three underlying factors captured children's responses: (1) fluid and speeded response (2) hypothesis testing and impulse control and (3) planning. The investigators interpreted their findings as evidence for stage-like development, with the first stage beginning around age 6, the next stage around age 10, and the final stage during adolescence. Six-year-olds, for example, were able to perform as well as adults on tasks that involved visual searching and planning simple sequences, whereas it was not until adolescence that the ability to plan complex sequences, verbal fluency, and motor planning reached maturity.

In contrast to the research on children's emerging capacities to regulate states and emotions, relatively little is known about how parents, other adults, and features of children's early environments affect the development of attention regulation and emerging executive functions. These influences clearly matter (Carlson et al., 1995), but researchers have yet to identify the mechanisms that account for individual differences among young children. Research on school-age children demonstrating that it is possible to teach attentional skills and executive functions to individuals with developmental disabilities (Borkowski and Burke, 1996; Graham and Harris, 1996) also indicates that they are amenable to environmental influence.

Deficits in Executive Function and Attention

The importance of understanding how children learn to plan and organize new actions, remember past experiences and bring them to bear on new experiences, and maintain attention to tasks is underscored by the consistent relation of deficits in any one of these processes to problems in school (Lyon, 1996). Of great interest to educators and parents alike is growing evidence that deficits in attention regulation and behavioral control are integral to disorders such as attention deficit hyperactivity disorder (ADHD). They also have vast implications for social and emotional behavior given the highly permeable boundaries between cognitive abilities and social competence. For example, forethought and planning are intimately involved in making friends, seeking attention, and solving interpersonal conflicts. Because social interactions involve people and people are often unpredictable and have their own goals that may interfere with one's own, flexibility is essential for achieving social goals (Goodnow, 1987). To be a competent social problem solver, one must be able to detect obstacles that will interfere with social goals, generate and evaluate alternative strategies to overcome or prevent these obstacles, and be able to flexibly adapt behavior to meet the challenges presented by the constantly changing social environment (Rubin and Krasnor, 1986). These are extremely challenging tasks for young children.

Designing appropriate, individualized interventions for young children who are displaying early deficits in organizational, planning, and attention-related capacities depends on understanding the processes that underlie their development and manifestation. Taking ADHD as an example, this is generally considered to be a relatively common disorder, with prevalence estimates ranging from 0 to 16.7 percent with a median of 2 percent for school-age children (Lahey et al., 1999). Yet the true prevalence and the cause of ADHD remain unknown (Zametkin and Ernst, 1999). According to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994), the diagnosis of ADHD requires evidence of either inattention or hyperactivity and impulsivity. There are three subtypes of ADHD, the predominantly hyperactive-impulsive type, the predominantly inattentive type, and the combined type. Most of the research to date is on the hyperactive-impulsive and combined types. Focus on the inattentive subtype is an emerging research topic.

Tests of the validity of the three DSM-IV subtests of AHDH for preschoolers have been conducted with children 4 to 6 years of age. The findings show that the subtests are valid for children of these ages (Lahey et al., 1998). Yet diagnosis is complicated by the fact that many of the behaviors associated with ADHD are normal for preschoolers who often have difficulty paying attention and are impulsive and fidgety by nature. And there is no information about the validity of ADHD diagnosis below age 4. The absence of a nationally representative epidemiological study of mental health problems in the childhood population leaves us in the dark with regard to efforts to distinguish children who are at the ends of a typical spectrum from those who are manifesting serious delays. A related critical issue in both research and practice concerns how to differentiate ADHD from other frequently coexisting conditions, such as learning disabilities, oppositional-defiant disorder, conduct disorders, and anxiety disorders and also how to individualize treatment when ADHD occurs in conjunction with these and other disorders. A recent study of 7- to 9-year-old children shows that different types of treatments for ADHD benefit individuals with different combinations of problems (MTA Cooperative Group, 1999). Finally, recalling the core concepts in Chapter 1, it is critically important to note that many of the symptoms of ADHD are also nonspecific indicators that something is not quite right. Heightened activity can also occur, for example, when a child is overly tired or upset. This can make the clinician's task of sorting out true disorder from other problems exceedingly difficult, perhaps especially with a diffusely defined disorder such as ADHD (Rutter and Sroufe, in press).

In sum, understanding the constructs of attention, memory, and executive function is critical for understanding how children think, learn, and develop. Because these constructs are hard to define and have overlapping boundaries, there is a pressing need to develop more refined definitions of executive function and its component skills, and for valid measures of early manifestations of pertinent behaviors and abilities (Weinberg et al., 1996). The capacity to map specific executive functions onto specific areas of the brain and to distinguish normal development from emerging disorders is dependent on such efforts to lend greater precision to the analysis of task demands and individual children's responses. For these important capacities, there is much to learn about the emergence, integration, and consolidation of the skills that serve as building blocks for adaptive behavior and provide a yardstick against which to evaluate what is abnormal neurological or cognitive development.

It is clear at this point that the various components of active, internally guided regulation of attention, behavior, and emotion emerge in intricately interrelated ways at the end of the first year of life and then develop more rapidly during the toddler and preschool years (Kopp, 1982; Rothbart and Bates, 1998). They emerge in the context of caregiving relationships that explicitly guide the child from her dependence on adults to regulate virtually every aspect of functioning to gradually taking over and self-regulating her own behaviors and feelings in one aspect of her life after another. Parents don't tell their 3-year-olds to clean up their rooms and then leave them to their own devices. They help by suggesting that they start with the clothes and throw them in the hamper and then put away their blocks on the shelf. They praise them at each step along the way and then move on to pulling the covers up on the bed. They repeat these patterns over and over, day after day, and then begin to pull back the scaffolding as the child begins to do it himself. Some children need more scaffolding than others and for longer times. Some children have serious disorders, such as mental retardation, that interfere with their ability to benefit from parents' and others' efforts to help them practice and then extend their emerging competencies. Sorting out children who don't have sufficient help from those with serious disorders is extremely challenging. In attempting to distinguish normal development from emerging disorders in these regulatory domains, it is crucially important to consider the contexts within which children's executive functioning and self-regulatory behaviors have been struggling to develop.